You can add a failed node back to the cluster if you have identified and fixed the issue that originally made the node unavailable and suitable for being marked as failed over.
When a node is marked as failed over, no changes are made on the failed node. The persisted data files on disk remain in place, however, the node will no longer be synchronized with the rest of the cluster. You cannot add a failed over node back to the cluster and re-synchronize. Instead, the node will be added back and treated as a new node.
Before adding a node back to the cluster, it is best practice to either move or delete the persisted data files before the node is added back to the cluster.
If you want to keep the files, you can copy or move the files to another location (for example another disk, or EBS volume). During the node addition and rebalance operation, the data files will be deleted, recreated and repopulated.
For more information on adding the node to the cluster and rebalancing, see Section 5.3.2, “Performing a Rebalance”.